RouteLLM is an open-source framework introduced in 2024 for optimizing Language Large Model (LLM) routing, aiming to reduce costs without degrading quality. It offers a drop-in replacement for OpenAI's client, and is compatible with multiple LLMs, making it highly efficient and cost-effective. The framework also facilitates easy integration of new routers and models, and provides in-depth instructions for calibration, evaluation, server setup, and configuration.